Re: [DotGNU]Re: [arch-users] In defense of (the idea of) having an XML-RPC api

[Stephen J. Turnbull]

>>>>> "Norbert" == Norbert Bollow <address@hidden> writes:

    Norbert> Stephen J. Turnbull <address@hidden> wrote:

> DotGNU is just as dangerous as .NET.  "The medium is the message",
> McLuhan taught us.  And what is the message of the medium?
> 
>                   THIS DATA YOURS NOW DATA OURS IS.

    Norbert> Addressing this issue has been part of the DotGNU vision
    Norbert> since the beginning.

DotGNU can only address this issue by providing multiple competing
vendors of most of the important webservices.  It cannot address it by
saying "don't do that" (the definition of _free software_ prohibits
that); it cannot address it by requiring provision of user access to
user data (the definition of free software prohibits that, too).

    Norbert> It's part of the definition of "DotGNU webservice" that
    Norbert> the owner of the data must be able to download the data
    Norbert> as well as the webservice program (that runs on the
    Norbert> server) which acts on this data.  [... W]e can use
    Norbert> trademark law to prevent them from using the term "DotGNU
    Norbert> webservice" when marketing ... "lock-in webservice
    Norbert> services".

Nobody with an excellent new algorithm will care about that.  What
they want is a distribution medium, which they will get from DotGNU.
Their users already know they're dealing with a monopolist (a la Bell
Labs/Karmarkar, who had the users send their data on tape by FedEx).
The important trademark will be that of the algorithm, not "DotGNU."

Of course an "algorithm" (maybe even patented) is an extreme case, but
I would not at all be surprised if DotGNU ends up ashamed to be
associated with a majority of the revenue generated by DotGNU-based
services.

    Norbert> It is my considered opinion that webservice vendors which
    Norbert> sell such "lock-in webservice services" will not be able
    Norbert> to compete successfully with vendors of DotGNU
    Norbert> webservices.  (I believe that most buyers of IT services
    Norbert> are wise enough to understand that lock-in situations are
    Norbert> not in their best interest).

Yeah, but they're also wise enough to recognize that being locked-in
to a service that is 20% more expensive but 50% more effective than
competing providers (including in-house provision) may be a good bet.
Who says there will be _any_ "competing DotGNU vendors"?  How are they
going to be capitalized?  Who says they will have services of
competitive quality, effectiveness, and reliability?

There will be _some_, of course.  But I'll bet you my Euros against
your yen that they concentrate in areas where there's already a lot of
competition, so that they don't care about throwing away lock-in if
they can reduce costs.

    >> To do good, DotGNU needs to concentrate on the _business_ side
    >> of offering _cheap_ _high-quality_ services to the bovines, to
    >> keep the herd out of the corrals of Gates and Jobs.  And the
    >> services have to be good enough to undercut internal corporate
    >> IT empires.

    Norbert> In the arena of "outsourceable webservices" this is what
    Norbert> DotGNU needs to be focused on, I fully agree with that.
    Norbert> Note bene, besides that there is another arena where
    Norbert> webservice protocols are important and relevant, namely
    Norbert> the arena of system integration.

This is more or less what I mean by "internal corporate IT empires".
Every sysadmin I know considers the one s/he works for to be a bigger
threat than Microsoft.  So you need to make outsourcing to DotGNU
vendors sufficiently robust and cheap to make CEOs and CFOs think
about going against the fierce opposition of the CIO.

I think there are plenty of good reasons to pursue the DotGNU program.
Trying to use DotGNU as a club to induce more use of free software is
just not going to work, though.  (IMO, and if I discover a technique
that changes my mind, I'll certainly let you know.  :-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.